CivArchive
    MOSS Photographic Realism Int8 - 1.0
    NSFW
    Preview 146889330
    Preview 146889364
    Preview 146889385
    Preview 146889486
    Preview 146889492
    Preview 146889520
    Preview 146889518
    Preview 146889522
    Preview 146889525
    Preview 146889526
    Preview 146889725
    Preview 146889723
    Preview 146889730
    Preview 146889728
    Preview 146889734
    Preview 146889939
    Preview 146889977
    Preview 146890090
    Preview 146891790
    Preview 146891428

    MOSS Photographic Realism Int8

    Hey everyone! 👋

    Today I’m excited to share my newly fine-tuned model: moss_Photographic_realism_Int8.

    I’ve spent hours tweaking and optimizing this model to deliver the best possible visual experience while maintaining an efficient footprint.

    🌟 Highlights & Features:

    • Photographic Lighting & Colors: Outstanding light rendering and tone balance that give your outputs a genuine, camera-shot feel straight out of the box.

    • No LoRAs Required: This model is fully self-sufficient—you don't need any additional LoRAs to achieve realistic results.

    • Plug & Play / Universal Workflow: No complex custom setup needed! You can simply use the standard default ComfyUI workflow for MOSS-Image.

    ⚠️ What to Expect (Pros & Cons):

    • Pros: Incredible atmospheric depth, warm natural lighting, realistic textures, and fast inference thanks to the INT8 quantization.

    • Cons (Known Limitations): Because of its lightweight parameters and high compression, minor artifacts or distortions can occasionally appear. This is completely normal for a model of this size, but the trade-off in speed and accessibility is well worth it!

    ⚙️ Recommended Settings:

    • Sampler: euler_sde / euler

    • Scheduler: simple

    • Steps: 25 - 30

    • CFG / Guidance: 2.5 - 3.5

    💡 Looking for even higher precision?

    If you prefer maximum fine detail with significantly fewer artifacts, feel free to check out my other release: Krea2_Realism_Fusion_(Int8) available right here on my Civitai profile! https://civarchive.com/models/3016926/krea2realismfusionint8

    Thank you all so much for your continued support! I hope you enjoy generating with this new model. As always, feel free to drop your renders and feedback in the gallery below! 📸✨

    Description

    Initial release of moss_Photographic_realism_Int8:

    - Fine-tuned & INT8 quantized for faster generation and lower VRAM usage.

    - Enhanced photographic lighting, color accuracy, and camera-like aesthetic.

    - Self-sufficient model: Works standalone without needing additional LoRAs.

    - Designed for native ComfyUI MOSS-Image workflows.

    - Recommended settings: 25-30 steps, Euler SDE / Euler sampler, simple scheduler, CFG 2.5-3.5.

    FAQ

    Checkpoint
    MOSS Image
    by kimrf
    Download (Beta) View on CivitAI
    realism
    base model
    nudity
    photography
    photorealism
    nsfw

    Details

    Downloads
    1,738
    Platform
    CivitAI
    Platform Status
    Available
    Created
    10/26/2026
    Updated
    11/6/2026
    Deleted
    -

    Files

    moss_Uncensored_image.json

    Size:
    38.15 KB
    SHA256:
    bc0a4cd317079139322884c96298037c188933f659c0f3973036e6818b098dc4

    Mirrors

    CivitAI (1 mirrors)

    flux2-vae_small.safetensors

    Size:
    237.96 MB
    SHA256:
    d8d52ba036475f5fb07c8b435e176d3d97ebfa82f0d1a1c317f9cc1e25bd013b

    Mirrors

    HuggingFace (30 mirrors)
    CivitAI (9 mirrors)
    ModelScope (1 mirrors)

    mossPhotographic_sftInt810_fp32.safetensors

    Size:
    237.96 MB
    SHA256:

    Mirrors

    moss_Photographic_realism_Int8.safetensors

    Size:
    5.04 GB
    SHA256:
    5996c6e426baa3441969998fab8bf2aa5eed79629f5ddd69d0befb7b957eff44

    Mirrors

    CivitAI (1 mirrors)

    mossPhotographic_sftInt810_3331776.safetensors

    Size:
    1.11 GB
    SHA256:
    4abb14e39c1a59018907581e609f43616ec6810126144ec2b1d2e6f8969628fe

    Mirrors

    HuggingFace (6 mirrors)
    CivitAI (3 mirrors)
    Civision (2 mirrors)
    ModelScope (2 mirrors)

    mossPhotographic_sftInt810_3330767.safetensors

    Size:
    237.96 MB
    SHA256:
    d8d52ba036475f5fb07c8b435e176d3d97ebfa82f0d1a1c317f9cc1e25bd013b

    Mirrors

    HuggingFace (30 mirrors)
    CivitAI (9 mirrors)
    ModelScope (1 mirrors)